Beauty in the blink of an eye: The time course of aesthetic experiences

Page created by Ken Barrett
 
CONTINUE READING
Beauty in the blink of an eye: The time course of aesthetic experiences
63

                                     British Journal of Psychology (2018), 109, 63–84
                                                © 2017 The British Psychological Society
                                                                                           www.wileyonlinelibrary.com

Beauty in the blink of an eye: The time course of
aesthetic experiences
San Verhavert, Johan Wagemans and M. Dorothee Augustin*
Laboratory of Experimental Psychology, University of Leuven (KU Leuven), Belgium

         Under normal circumstances, perception runs very fast and seemingly automatic. In just a
         few ms, we go from sensory features to perceiving objects. This fast time course does not
         only apply to general perceptual aspects but also to what we call higher-level judgements.
         Inspired by the study on ‘very first impressions’ by Bar, Neta, and Linz (2006, Emotion, 6,
         269) the current research examined the speed and time course of three aspects of the
         aesthetic experience, namely beauty, specialness, and impressiveness. Participants were
         presented with 54 reproductions of paintings that covered a wide variety of artistic styles
         and contents. Presentation times were 10, 50, 100 and 500 ms in Experiment 1 and 20, 30
         and 40 ms in Experiment 2. Our results not only show that consistent aesthetic
         judgements can be formed based on very brief glances of information, but that this speed
         of aesthetic impression formation also differs between different aesthetic judgements.
         Apparently, impressiveness judgements require longer exposure times than impressions
         of beauty or specialness. The results provide important evidence for our understanding of
         the time course of aesthetic experiences.

From the moment we open our eyes, we clearly see the world around us and quickly
extract information and meaning from it. Studies have found that when presented with
real-world images, people are able to detect objects based on presentations as short as
50 ms and to recognize objects after only about 100 ms of presentation (see, e.g., Fei-Fei,
Iyer, Koch, & Perona, 2007; Grill-Spector & Kanwisher, 2005). However, this fast
processing comprises different stages. For example, Fei-Fei et al. (2007) found that before
40 ms, sensory-related features (light–dark) are dominant for people’s perception of
stimuli, while from around 50 ms of presentation, a shift to more object-related features
occurs. Subjects are able to name very general object categories like manmade objects
(‘furniture’, ‘desk’) and gradually get more detailed and accurate over time (Fei-Fei et al.,
2007). Such a coarse-to-fine development of percepts has been found in many
psychophysical studies (for a review, see Hegde, 2008) and is a core characteristic of
the percept formation or microgenesis (Bachmann, 2000). What subjects perceive
increases in detail the longer they see it, even if the image does not change. This happens
on a very fast timescale (Fei-Fei et al., 2007; Hegde, 2008). A very vivid illustration that
such speed and differentiated time course do not pertain only to general perceptual and
identification aspects, such as colour or object category, but also to judgemental aspects

*Correspondence should be addressed to M. Dorothee Augustin, Laboratory of Experimental Psychology, KU Leuven, Tiensestraat
102, box 3711, 3000 Leuven, Belgium (email: mdorothee.augustin@posteo.de).
San Verhavert has moved since this study was undertaken and is now based at Department of Training and Educational Sciences,
Research group Edubron, University of Antwerp, Gratiekapelstraat 10, 2000 Antwerp, Belgium.
M. Dorothee Augustin has moved since this study was undertaken and is now based in Munich, Germany.

DOI:10.1111/bjop.12258
64   San Verhavert et al.

of person perception, was presented by Bar, Neta, and Linz (2006; for further work on this
topic, see, e.g., Todorov, Pakrashi, & Oosterhof, 2009 or Todorov & Porter, 2014).
     In their study, Bar et al. (2006) examined whether people were able to form consistent
‘first impressions’ about others and how fast they could do this, that is, what minimal
exposure time allowed them to come to consistent person judgements. Participants saw
photographs of faces at 26, 39 and 1,700 ms, which they had to rate on an emotion-related
feature (threateningness) and a personality-related feature (intelligence). Because such
judgements – despite important influences of simple physical features for social trait
impressions (Vernon, Sutherland, Young, & Hartley, 2014) – are inherently subjective,
‘correctness’ of judgements was determined as judgement consistency over time, by
calculating the correlation between judgements with short exposure times (26 and
39 ms) on the one hand and long exposure time (1,700 ms) on the other hand. It appeared
that people were consistent in judging the emotion-related feature, that is, threatening-
ness, from whatever they could perceive within 39 ms – but this was not the case for the
personality-related measure, intelligence. These results suggest that not only percepts per
se develop in time – but also our judgements of these percepts (see also Bachmann &
Vipper, 1983). To what extent these findings can be extended to other impressions and
other aspects of impression formation remains to be investigated. The current study taps
into this issue with a focus on the field of aesthetics: How much viewing time do people
need to form aesthetic impressions, and to what extent does that differ between different
aesthetic judgements?
     Aesthetics has more and more become a ‘hot topic’ in psychological literature in the
past few years (see, e.g., Augustin & Wagemans, 2012; Jacobsen, 2010; Leder & Nadal,
2014; Redies, 2015). The exact nature of the processes involved in attraction, rejection,
etc., is still relatively unclear, although it is clear that aesthetic experiences involve a
complex interplay of cognitive and affective processes assessed by different aesthetic
judgements (Leder, Belke, Oeberst, & Augustin, 2004). Moreover, models of aesthetic
experiences (Leder et al., 2004; Leder & Nadal, 2014; for a comparable model regarding
music, see Brattico, Bogert, & Jacobsen, 2013) clarify that aesthetics- and art-related
perception are also subject to an underlying time course. Leder et al. (2004) define the
aesthetic experience as a process to ‘cognitively master [an] artwork’ (p. 493), which is
accompanied by continuously upgrading affective states and results in ‘an (aesthetic)
emotion’. (p. 493). The person’s affective state can be assessed at any of these stages and
may differ depending on the state of processing. Thus, the model conceives of aesthetic
experiences as complex phenomena, in which gradual development and change are
inherent.
     Early evidence favouring this assumption came from Cupchik and Berlyne (1979).
They explored how fast people could discriminate collative properties of paintings
and whether these collative properties had an effect on perceived pleasingness.
Collative properties refer to the relationship between elements within the stimulus
and include the degree of ambiguity, complexity, and familiarity (Berlyne, 1971;
Cupchik & Berlyne, 1979). The results of the Cupchik and Berlyne (1979) study
showed that people are able to discriminate collative properties (and paintings
varying on those properties) very quickly, with presentation times (PTs) of 50 ms
being enough to gather the relevant visual information. According to the authors, this
indicates that the initial stage of visual processing is holistic and is a stage of higher
alertness and tension.
     In 1983, Bachmann and Vipper replicated and extended the findings by Cupchik and
Berlyne (1979). Using rating scales derived from Berlyne’s collative scales (Berlyne, 1974),
Beauty in the blink of an eye   65

they found that people could not only differentiate art styles based on collative properties,
but that this differentiation already seemed to be apparent starting from a presentation
time of 1 ms only. It has to be taken into account that trials were unmasked, which makes
interpretations in terms of exact timing difficult. Nevertheless, the results by Bachmann
and Vipper (1983) suggest that people only need very brief glimpses of visual information
to differentiate different kinds of art.
    Locher, Krupinski, Mello-Thoms, and Nodine (2007) took a more general approach to
the time course of art perception to find out in what order different aspects of artworks are
perceived. Based on their subjects’ free reports and eye movements, they propose that the
change in pictorial properties of a percept (e.g., symmetry, complexity, structural
features) might already reach a highly advanced stage after about 100 ms: Within 100 ms,
people can extract enough information to form a significant holistic impression of the
semantic meaning (i.e., gist) of paintings, including expressive aspects and ‘meaningful’
aesthetic judgements, as indicated by a significant correlation between pleasingness
ratings at 100 ms and at unlimited PT. The observation that people can form a holistic
impression after a single glance (100 ms; Locher et al., 2007) is in line with the finding
that people can describe a scene in superordinate categories at similar speed (after 107 ms
of presentation, according to Fei-Fei et al., 2007; 50 ms, Grill-Spector & Kanwisher,
2005). Extending this work to more specific properties of artworks, Augustin, Leder,
Hutzler, and Carbon (2008) found that processing of content in art can already be traced
after 10 ms glances and is already strongly developed after presentations of around 50 ms,
whereas style processing emerges from about 50 ms of presentation and develops more
slowly. Thus, content processing seems to be faster than style processing, but the latter
is – nevertheless – impressively fast if one thinks of the complexity one would normally
assume for style judgements.
    The studies cited above – implicitly or explicitly – all work with a ‘classical’
microgenetic approach (see Bachmann, 2000) that systematically varies perceptual
conditions and thus the informational basis at each processing stage. Others have
examined temporal aspects of art perception and aesthetic phenomena via alternative
methods, esp. EEG, to examine the duration of mental processes (e.g., Augustin,
Defranceschi, Fuchs, Carbon, & Hutzler, 2011b; Jacobsen & H€       ofel, 2003). For example,
Jacobsen and H€   ofel (2003) presented evidence that the brain seems to differentiate
between beautiful and non-beautiful patterns in a time window of about 300 and 400 ms
post-stimulus onset, as reflected in an early frontocentral phasic negativity specific to non-
beautiful judgements.
    The previous studies all illustrate that there is a time course in the perception of
relational (‘collative’) properties of the percept of paintings (Bachmann & Vipper, 1983)
as well as in the perception of style and content (Augustin et al., 2008). It has also been
shown that the percept and differences and/or changes in the percept influence
perceived arousal and pleasure (Cupchik & Berlyne, 1979; Locher et al., 2007). In
addition, studies focusing on mental chronometry have rendered first estimations of the
duration of processing underlying beauty judgements (Jacobsen & H€      ofel, 2003). Very little
is yet known about the temporal unfolding of the aesthetic experience in terms of the
development of its different facets. As discussed above, aesthetic experiences are
supposed to be very complex in nature, comprising different kinds of aesthetic
judgements and emotions at different stages of processing (Brattico et al., 2013; Leder
et al., 2004), with beauty only being one – even though very important – facet of many
(Augustin, Wagemans, & Carbon, 2012; Jacobsen, Buchta, K€          ohler, & Schr€ oger, 2004;
Knoop, Wagner, Jacobsen, & Menninghaus, 2016). To find out more about the temporal
66   San Verhavert et al.

unfolding of different aesthetic judgements will thus be an additional step to shed further
light on the processes involved in aesthetic experiences, which – as stated above – still
remain rather poorly understood, in spite of a growing body of systematic research.
    The current study adapted the logic followed by Bar et al. (2006) to the realm of art
perception, in order to find out how quickly different aesthetic impressions can be
formed, that is, how short the presentation time can be for people to come to different
aesthetic judgements, and to what extent this time course of aesthetic impression
formation differs between different aesthetic judgements. We also examined differences
in response time (RT) between judgements as an indicator of judgement complexity. Our
study follows a ‘classical’ microgenetic approach (e.g., Augustin et al., 2008; Bachmann &
Vipper, 1983): Viewing conditions are systematically varied in order to find out about the
stages involved in the formation of aesthetic judgements.
    Based on the literature and our previous studies, we focused on three aesthetic scales:
beauty, specialness, and impressiveness.
    In empirical studies, beauty has emerged as a core concept of aesthetic experiences.
This could be shown not only in general (Jacobsen et al., 2004), but also when
comparing different visual object classes (Augustin et al., 2012) as well as, very recently,
in the field of literature (Knoop et al., 2016). In his factor-analytic studies, Markovic
(2010) found a factor ‘aesthetic experience’ that is dominated by aspects of specialness.
In our view, this is not a contradiction: Cupchik and Gebotys (1988) identified pleasure
and interest as two dimensions of aesthetic responses, with pleasure being more
affectively and interest more cognitively based. In the Augustin et al. study (2012),
terminology around the idea of ‘being special’ played a central role for visual art (besides
beauty), and beauty, specialness, emotiveness, and impressiveness were important
aesthetic factors in another study by the same authors (Augustin, Carbon, & Wagemans,
2011a). Our pilots for the current research project, however, showed that ‘emotional’
(‘emotioneel’ in Dutch) showed very little variance. This may be due to the fact that
subjects in our study viewed reproductions of paintings on a computer screen (for details
of method, see further) and that judgements of emotionality may be much more sensitive
to circumstances than other aesthetic judgements. To avoid such biases in the results, we
replaced this scale by impressiveness. ‘Impressive’ also appeared as an important
aesthetic term, especially for landscapes and buildings, in the study by Augustin et al.
(2012). The concept of impressiveness shows close relatedness to the idea of the
sublime, which holds a special relationship with beauty and the aesthetic in philosophy
(e.g., Burke, 1990; originally published in 1757; Kant, 1991; originally published in 1764;
Schopenhauer, 1948, originally published in 1818). According to Burke (1990, originally
published in 1757), the sublime and the beautiful constitute opposite poles, with the
former being related to potentially dangerous or terrifying objects and the latter to
pleasing, ‘nice’ objects and tension relief.
    The current study aimed to uncover the time course of these three scales. Based on the
literature, we chose presentation times of 10, 50, 100, and 500 ms in a first step and 20, 30,
and 40 ms in a second step. According to Augustin et al. (2008), content processing starts
with PTs of around 10 ms and is already strongly developed after 50 ms of presentation –
the point where a shift from sensory-related to object-related processing seems to occur
(Fei-Fei et al., 2007) and where people start detecting objects (Fei-Fei et al., 2007; Grill-
Spector & Kanwisher, 2005). The same study by Augustin et al. (2008) suggests that style
processing supposedly emerges around 50 ms of presentation. Another important time
point in visual processing seems to lie around 100 ms until which subjects are able to
perceive the global scene but not the details (Fei-Fei et al., 2007). This also seems to be the
Beauty in the blink of an eye        67

point where subjects can form a holistic impression of a painting and can make consistent
judgements about it (Locher et al., 2007). As our pilot studies showed that there is already
a strong development in aesthetic judgements between the 10 and 50 ms conditions, we
decided to conduct a second experiment to scrutinize this time frame. In both
experiments, unlimited presentation times served as the baseline. Following the logic
presented by Bar et al. (2006) our main measure were the correlations between the
judgements at each of the limited PTs (10, 20, 30, 40, 50, 100, and 500 ms) and at
unlimited viewing time. These were regarded to indicate how far developed, that is, how
consistent, a judgement is at different presentation durations. In sum, we conducted two
experiments: a first with 10, 50, 100 and 500 ms PTs and an unlimited presentation
baseline, and a second with PTs of 20, 30, 40 ms, and unlimited presentation baseline.

Method
Pre-study
The pre-study served to select diverse materials with a broad range in both style and
content that allowed for a wide range of aesthetic impressions. Starting point was a list of
18 Western European and North American art styles, drawn together in collaboration with
an art historian. The selection of art styles was made on the basis of commonness, as
measured by their appearance in art books and museums. In addition, the styles were to
represent a wide range of styles, with as little overlap as possible between them. For each
art style, we chose between three and 15 paintings of diverse style and content, and, for
the sake of diversity, by at least three different painters. This procedure resulted in a basic
set of 225 paintings. Reproductions were downloaded from the DVD 25,000 Meisterw-
erke [masterworks] (Directmedia Publishing: zeno.org, 2007), www.artstor.com (Art-
stor), from sites of museums listed on www.artcyclopedia.com (Artcyclopedia.com) and
from Wikipedia. Only high-quality reproductions of paintings were accepted. All pictures
were resized to 140,000 square pixels at 72dpi.
    Twelve university students participated (10 women; M(age) = 20.38 years, SD
(age) = 1.80 years). None of them had received any formal training in art or art history
beyond regular school education.
    Participants judged the paintings in two blocks, during each of which all paintings
were presented in random order at unlimited presentation time. In the first block,
participants were asked to rate the strength of their aesthetic impression on a scale from 1
to 7, with 1 = a very weak impression and 7 = a very strong impression – positive or
negative. This formulation was used to be sure people rated the strength and not the
valence of their impression. In the second block, participants were asked whether they
had seen the paintings before they participated (‘yes’/’no’).
    Paintings known by 10% of participants or more were removed from the analyses. To
make it easier to compare ratings of the strength of aesthetic impression (M = 3.460,
SD = 1.601) between paintings, z-scores were calculated per painting. On the basis of
these z-scores, three paintings were selected per art style, one that made a weak
impression (z  1), one that made an intermediate impression (z  0), and one that
made a strong impression (z  1).1 This resulted in a selection of 54 paintings for the main
study (see Appendix A).

1
  If more reproductions met the z-score criteria, two additional criteria were used: (1) a broad range in style and content and (2)
three different painters within every style.
68   San Verhavert et al.

EXPERIMENT 1
Participants
Twelve paid volunteers participated in this experiment. Our sample consisted of four
male and eight female participants aged 18–57 (M = 29.8 years, SD = 13.0 years). Three
participants were working, and the others were students. None of them had received any
formal training in art or art history beyond regular school education. All had normal or
corrected-to-normal vision.

Stimuli
The stimuli consisted of 54 reproductions of paintings of 18 different art styles from the
15th century to the 1960s (see Appendix A), selected in a pre-study, as described above.
Figure 1 shows black-and-white versions of two artworks used for the study.
    These stimuli were also used to construct coloured noise masks via an algorithm that
has been proven useful to yield efficient masks (Torfs, Panis, Barthlema, & Wagemans,
2012). The algorithm divided our stimuli into squares of 5 9 5 pixels, removed all
completely white squares from this pool, and randomly assembled the squares to form
new combinations of 561 9 497 pixels. Eighteen of such combinations were created to
serve as masks.

Procedure
The experiment was programmed and controlled by the experimental software E-Prime
(Psychology Software Tools Inc.) and run on a Dell Precision T5400 PC. The screen had
a resolution of 1,280 9 1,024 and a 100-Hz refresh rate. The stimuli sizes ranged from
10.37 to 18.77 cm in width and from 9.24 to 16.79 cm in height. With a viewing
distance of about 60 cm, this resulted in visual angles ranging from 9.06° to 16.10° in
the horizontal and from 8.09° to 14.48° in the vertical axis. All instructions were given
in Dutch.

Figure 1. Two black-and-white versions of artworks used for the study (left to right): ‘The Alexander
battle (Battle of Issus)’ by Albrecht Altdorfer, 1529. © Bayerische Staatsgem€aldesammlungen – Alte
Pinakothek Munich. ‘Pathway in a field’, by Edgar Degas, 1890. © Yale University Art Gallery.
Beauty in the blink of an eye   69

    In a first block, participants were presented with ten practice trials that were identical
to the experimental trials except for the stimuli (see Appendix B). The experiment was
divided into two blocks. In the first block, stimuli were presented at 10, 50, 100, and
500 ms. The stimuli and stimulus–PT combinations were randomly intermingled. In the
second block, the same stimuli were presented at unlimited PT. Each trial consisted of a
250-ms presentation of a fixation cross, followed by a stimulus at one of the four PTs and a
250-ms presentation of a coloured noise mask (Figure 2a). Then, ratings on three 7-point
Likert type of scales were asked for all the scales at the same time, beautiful (‘mooi’ in
Dutch), special (‘speciaal’ in Dutch), and impressive (‘indrukwekkend’ in Dutch)
(1 = ‘not at all’ to 7 = ‘very much’). The order of the scales was balanced over participants
using a Latin square logic. In order not to confuse participants, the order of scales was
identical within participants in both blocks.
    At the end of the experiment, participants were asked whether they had seen the
paintings before as well as some questions concerning their experience with art, namely
whether they studied or had taken courses in art history, how many times they visited a
museum, and how many art books they possessed. Participants were also asked whether
they used any strategies in their ratings. This information allowed us to get an idea about
potential artefacts in the results.
    In total, participants gave 54 (paintings) 9 5 (presentation times) 9 3 (scales) = 810
ratings.

EXPERIMENT 2
Experiment 2 had the same logic as Experiment 1, but shorter PTs were used, which led to
minor changes in the experimental procedure.

Participants
Twelve people (seven female and five male) participated in this experiment – none of
them had participated in Experiment 1 or the pre-study. The participants’ age ranged
between 21 and 55 years (M = 31 years, SD = 13 years). Three participants studied
psychology, two studied law, and one studied applied economical sciences. Five
participants were graduated and had a job, and one was unemployed. All participants

Figure 2. Visualization of the trial structure in (a) Experiment 1 and (b) Experiment 2. [Colour figure can
be viewed at wileyonlinelibrary.com]
70   San Verhavert et al.

were volunteers. None of the participants had received any formal training in art of art
history. The participants had normal or corrected-to-normal vision.

Stimuli
Stimuli and the coloured noise masks were the same as in Experiment 1.

Procedure
The stimuli were presented at 20, 30, and 40 ms and unlimited PT. Because of the
shortness of the PTs, we assumed it might be too difficult for participants to give all three
judgements in a row (assuming that quickly flashed pictures also leave only short-lived
visual memories). For this reason, it was decided to include only one scale per block for the
short PTs. The order of blocks was completely balanced across participants. A fourth
block was included for unlimited presentation time. The fourth block was also divided
into three sub-blocks, one per scale. For the sake of clarity, the order of sub-blocks was the
same within participants as the order of the first three blocks. To familiarize participants
with the experiment, there were two practice blocks, each with a different scale. Apart
from this, the practice trials were identical to those of blocks 1 and 2 in Experiment 1
(Figure 2b) for limited and unlimited PT, respectively.
    In total, participants gave 54 (paintings) 9 4 (PTs) 9 3 (scales) = 648 ratings.
    At the end of Experiment 2, we asked the same questions on familiarity and art
knowledge as described for Experiment 1.

Results
General descriptive results and data preparation
Before analysing the results, the data were inspected for possible outliers. On the basis
of the RT histograms, we defined a cut-off value of 10s. This caused 0.6% of the data for
Experiment 1 and 0.5% for Experiment 2 to be excluded from further analysis.
    The remaining data were used to answer the following questions: What is the minimal
presentation time that people need to extract the relevant information for different
aesthetic judgements? Do different aesthetic judgements develop differently over viewing
time? And how fast or slow are people in making these different judgements?
    In order to directly compare differences between the scales, the type of judgement
(Scale) was included as a variable in the ANOVAs rather than considering the scales as
different dependent variables. In addition, it can be assumed that the scales used are
aspects of a broader aesthetic experience and are thus not fully separate measures.
    In all our repeated-measures ANOVAs, p-values were obtained using Greenhouse–
Geisser ^e corrected degrees of freedom (df).
    Figure 3a,b plots the means of the ratings per scale per PT, with 1 SD error bars.

Onset and development of aesthetic judgements
To see how the aesthetic judgements develop in time, judgements at limited presentation
times were correlated with the corresponding judgements at unlimited PT. We assume
that the judgement at unlimited PT is based on a comparatively advanced aesthetic
impression (assuming that a ‘full’ aesthetic impression can perhaps never be assessed, at
Beauty in the blink of an eye   71

Figure 3. Means and one standard deviation error bars of Judgement (y-axis) per Presentation Time
(PT; x-axis) for (a) Experiment 1 and (b) Experiment 2, separated by Scale.

least not under experimental circumstances). The correlation between judgements at
limited and at unlimited PT then reflects how far the judgement is already developed after
a certain amount of viewing time (see also Introduction).
    To increase the power of the correlation analysis, we averaged over subjects and
calculated the Pearson correlations over stimuli (n = 54).

EXPERIMENT 1
None of the correlations was significant at 10 ms (ps > .05). Correlations for all the scales
were significant for presentation times of 50 ms and higher (ps < .05). Figure 4a shows
the correlations plotted against time.
72   San Verhavert et al.

Figure 4. Correlations between short PT and unlimited time (y-axis) per PT (x-axis) and one standard
deviation error bars divided by Scale for (a) Experiment 1 and (b) Experiment 2.

    To test for time-related trends, we performed a 3 (Scale) 9 4 (PT) repeated-measures
ANOVA and simple main effects analyses. In this case, Fisher’s z corrected correlations
were calculated per stimulus over persons (Fz(r)), and the ANOVA was conducted over
stimuli to increase its power.
    The ANOVA showed a main effect for PT, F(2.581, 136.781) = 32.017, p ≤ .0001,
g2 = .377, a main effect for Scale, F(1.940, 102.827) = 29.597, p ≤ .0001, g2 = .358,
and no interaction, F(4.840, 256.518) = 1.816, p = .112. More specifically, the simple
main effects analysis (for details, see Table C1 in Appendix C) yielded significant
differences between all PTs (ps ≤ .045), with higher PTs having higher correlations.
These results show that in the time window between 10 and 100 ms of presentation,
there is a general development of rating values over time. Furthermore, one can
observe a significant difference between the scales (ps ≤ .006). Impressive has lower
correlations than special, which again has lower correlations than beautiful. From this,
it can be concluded that overall beautiful develops faster than special, which appears
to develop faster than impressive. However, these time course differences between
scales appear to be general offset differences, not slope differences, as the Scale–PT
interaction is not significant.

EXPERIMENT 2
Experiment 2 examined the development of aesthetic impressions in more detail within
the window of viewing times between 10 and 50 ms, at 20, 30, and 40 ms. Here, the
correlations with unlimited PT (performed over stimuli) for beautiful and special were
not significant at 20 ms (ps > .05). Correlations were significant from 30 ms on
(ps < .05). The correlations for the scale impressive were not significant at any PT.
Figure 4b presents a plot of the correlations.
   To statistically compare the development of our three aesthetic judgements
between 20 and 40 ms of presentation, we conducted a 3 (Scale) 9 3(PT) repeated-
measures ANOVA with the Fz(r) of the judgements as dependent measure and simple
main effects analyses on the Fz(r). As in Experiment 1, the ANOVA was conducted
over stimuli, with Fisher’s z corrected correlations calculated per stimulus over
persons.
Beauty in the blink of an eye   73

Figure 5. Response Times (RT; y-axis) per PT (x-axis) and one standard deviation error bars divided by
Scale for (a) Experiment 1 and (b) Experiment 2.

    The ANOVA yielded a significant main effect of PT, F(1.891, 100.198) = 23.067,
p ≤ .0001, g2 = .303. We also observed a significant main effect of Scale, F(1.855,
98.318) = 8.447, p = .001, g2 = .137, but no interaction, F(3.642, 193.033) = .758,
p = .542. Simple main effects analyses (for details, see Table C2 in Appendix C)
showed significant differences between 20 and 30 ms, and between 20 and 40 ms
(ps ≤ .0001). The correlations at 30 and 40 ms were both higher than at 20 ms,
showing a strong development between the 10 and 50 ms conditions. However, the
difference between 30 and 40 ms is not significant (p = .417). Correlations for
impressive differed significantly from both beautiful and special (ps ≤ .0001), with
correlations for impressive being lower than for the other two judgements. The
74   San Verhavert et al.

difference between beautiful and special also reached significance (p = .041). The
correlation for beautiful was higher than for special. This suggests that between 10 and
50 ms of presentation impressive develops more slowly than both beautiful and
special, while special seems to lie somewhere in between impressive and beautiful in
terms of speed.

Speed of judgement
Differences in speed of judgements between the scales were analysed to find further hints
regarding the complexity or difficulty in the different judgements (see, e.g., Dodonov &
Dodonova, 2012). As participants were asked to answer as fast as possible, it is assumed
that the difficulty in judgement and the PT (amount of visual information) might lead to
variation in response times (RTs).
    The means of RTs ranged from 920 to 2,154 ms for Experiment 1, with standard
deviations between 195 ms and 470 ms (Figure 5a). In Experiment 2, the means of the
RTs lay between 1,180 and 2,669 ms and standard deviations ranged from 320 to 564 ms
(Figure 5b).

EXPERIMENT 1
The 3 (Scale) 9 4 (PT) repeated-measures ANOVA over stimuli with RT as dependent
measure yielded a significant main effect for PT, F(3.462, 1833.506) = 155.395,
p ≤ .0001, g2 = .746, and a significant main effect for Scale, F(1.805, 95.670) = 30.243,
p ≤ .0001, g2 = .363. Also a significant interaction, F(6.614, 350.530) = 8.187, p ≤
.0001, g2 = .134, was found. Simple main effects analyses (for details, see Table D1 in
Appendix D) showed that all differences in RT between PTs are significant (ps ≤ .012)
except between 50 and 100 ms and between 100 and 500 ms. RT increases with longer
PT. There is a significant difference between the scales impressive and beautiful and
between the scales impressive and special (p ≤ .0001). Impressive has longer RTs than
the scales beautiful and special. These differences are significant from PTs of 50 ms on
(ps ≤ .003). However, the difference between impressive and beautiful fails to reach
significance at unlimited PT. The difference in RT between beautiful and special is also
significant at 50 ms (p = .002) and at 100 ms (p = .003) with special having longer RTs.
    First of all, we observed longer RTs at longer PTs. This probably reflects a self-paced
responding that is in synch with shorter or longer PTs. However, the non-significant
difference in RT from 100 ms on can reflect a ceiling of time needed to process the stimuli
for their aesthetic value.
    In addition, it seems that from 50 ms on impressive ratings need a longer time than
ratings for beautiful and special, with the difference disappearing at unlimited PT. This
may potentially indicate that impressive is more complex and/or rational/cognitive than
beautiful and special, requiring more reflection.

EXPERIMENT 2
The 3 (Scale) 9 4 (PT) repeated-measures ANOVA over stimuli with RT as dependent
variable showed a significant main effect of PT, F(2.768, 146.680) = 344.940, p ≤ .0001,
g2 = .867, no main effect of Scale, F(1.933, 102.464) = 2.849, p = .064, and no
significant interaction, F(5.206, 275.904) = 1.813, p = .107. The simple main effects
analyses (for details, see Table D2 in Appendix D) yield only a significant difference
Beauty in the blink of an eye   75

between the short PT and unlimited PT (ps ≤ 0001). Possibly, a 20-ms interval is not long
enough to observe significant differences in processing length.

Potential effects of repeated exposure
Apart from our main questions, we checked for possible methodological effects in
Experiment 2. Does repeated presentation of the paintings have an effect on the
judgement? For that, the judgements were grouped according to the block the judgement
was given in. Pearson correlations between the judgements at limited time and unlimited
PT were calculated per block.
    A 3 (Block) 9 3 (PT) repeated-measures ANOVA on Fz(r) showed no significant effect
of block and thus no reason to worry about possible effects of the blockwise rating method
used in Experiment 2 (for ANOVA results, see Appendix E).

Discussion
The current study investigated the time course of aesthetic judgements, with a focus on
three impressions that have been shown to be important for experiences of visual art
(Augustin et al., 2012): beautiful, special, and impressive. We were interested in two
questions: What is the minimal presentation time on the basis of which people are able to
form stable aesthetic judgements? And to what extent do the time courses of the three
judgements differ – regarding both required viewing time and response times?
    First of all, our results illustrate that people are extremely fast at forming aesthetic
impressions. In many cases, a presentation of about 30 ms seems enough to extract the
information that is relevant to come to a meaningful aesthetic judgement – not more time
than is needed to extract the information pertaining to a face’s threateningness (Bar et al.,
2006). Such extreme speed is in line with similar results from the aesthetics literature (e.g.,
Bachmann & Vipper, 1983; Cupchik & Berlyne, 1979). For example, Cupchik and Berlyne
(1979) showed that participants differentiate paintings on the basis of the so-called
collative properties after no more than 50 ms of presentation. Augustin et al. (2008)
report first effects of content and style on similarity ratings of paintings after presentation
times of 10 and 50 ms, respectively. Such results suggest that the extraction of
information for ‘basic’ aesthetic processing does not necessarily require more time than
for other perceptual tasks – even though aesthetic encounters undoubtedly have a special
role for many people and aesthetic experiences stand out from other experiences in real
life (see also Allesch, 2006). Correlations with unlimited PT yet do increase over time, as
reflected in a main effect of PT in both experiments. This can be explained by reduced
uncertainty of the percept (Cupchik & Berlyne, 1979). Still, only a glimpse of an artwork
seems to be necessary to form an aesthetic judgement that is at least meaningful.
    How fast aesthetic impression formation actually is yet seems to depend on the kind of
judgement asked for: Whereas we find significant correlations with unlimited time from
30 ms of presentation on for beautiful and special, correlations for impressive only
become significant for PTs of 50 ms and longer. ANOVAs in both Experiment 1 and
Experiment 2 yield a significant main effect of judgement: Correlations with unlimited
time for impressive are overall lower than for the other two judgements, with special lying
in between beautiful and impressive. Thus, beauty seems to be the fastest aesthetic
impression of the three, impressiveness the slowest. Differences in PT required for a
meaningful judgement may, in turn, also imply that people need different kinds of visual
76   San Verhavert et al.

information in order to form the three judgements, because a specific kind of information
needed for a specific judgement can take a different amount of time in order to become
represented in the perceptual system.
    A similar picture is yielded by the inspection of response times, even though only in
Experiment 1 (from 50 ms on) and with a slightly less clear pattern: Participants not only
need shorter PTs to come to meaningful judgements of beauty and specialness than for
impressiveness; response times are significantly faster too. There is no general response
time difference between beautiful and special. The difference between beauty and
impressiveness vanishes at unlimited time. Still, if we assume a direct relation between
response time and processing time or task complexity (Dodonov & Dodonova, 2012),
impressiveness needs longest processing and may be the most complex or difficult of the
three judgements examined, despite potential confounds with motor execution times
(see, e.g., Augustin et al., 2011b).
    Taken together, it looks as if people only need very little time to extract the most
important information for judgements of beauty and specialness. Possibly, low-level
sensory visual information, which according to Fei-Fei et al. (2007) dominates perception
up to approximately 40 ms, may already be sufficient in this case. People can also detect
and categorize based on what they perceive within 30 ms (Grill-Spector & Kanwisher,
2005) – but they cannot consciously or explicitly name what they saw (Fei-Fei et al.,
2007). This supposedly changes after about 50 ms of presentation: According to Fei-Fei
et al. (2007), people then utter first descriptions/rough classifications of the objects they
see. Similarly, Augustin et al. (2008) found first effects of style on judgements from 50 ms
of presentation onwards, and Cupchik and Berlyne (1979) report the same time window
for extraction of collative properties of paintings (Cupchik & Berlyne, 1979). Thus,
explicit content classification, stylistic aspects, and collative properties are all among the
‘candidate’ information that may be of particular relevance for judgements of
impressiveness. Which of these kinds of information is most relevant is an important
question for further research. Moreover, this may differ between different kinds of
stimulus materials. Roye, H€  ofel, and Jacobsen (2008) even reported temporal differences
in the processing of facial beauty depending on the gender of the face. One potential
explanation offered by the authors is that people use less cues to evaluate male beauty
than female beauty. As for artworks, a relevant aspect to investigate in this respect is a
potential difference between abstract and representational artworks. It may be reasonable
to assume that content aspects and/or semantics are of particular importance for the
development of aesthetic judgements. As we only had 11 fully abstract works in our
stimulus sample, we are unable to provide a reliable analysis with our data, but the style of
paintings in general and abstractness in particular is a factor that requires investigation in
future research.
    What do the current results imply for our understanding of art perception? As
already mentioned above, it is beyond question that in order to experience an artwork
in all its flavour one needs time, freedom, an adequate setting (e.g., museum), and,
possibly, mindset (Leder et al., 2004; Wagemans, 2011). In a real-life museum, setting
viewing times will probably always exceed 10, 50, and also 500 ms (for a study on
viewing times in the museum, see Smith & Smith, 2001). Nevertheless, the results of the
current study reveal that people only need brief glimpses to form aesthetic judgements
that – at least according to what one can measure in the laboratory – are already
meaningful. Actually, the basics of aesthetic judgements already seem to be laid from
about 30 ms on.
Beauty in the blink of an eye   77

    In their 2012 study on word usage and underlying concepts in visual aesthetics,
Augustin et al. (2012) found that people are extremely differentiated in their aesthetic
word usage, especially with a view to different object classes. Beautiful is the universal
aesthetic term (see also Jacobsen et al., 2004), while special is mostly relevant for visual
art and clothing and impressive for visual art, landscapes, and buildings. Thus, different
object classes seem to ‘trigger’ different kinds of aesthetic impressions, which, in turn,
gives a hint at the nature of the different underlying aesthetic experiences. The current
research now shows that people are not only differentiated in their aesthetic judgements,
but that these judgements also differ in their general time course – even within object
class. For visual art, beauty is fastest, impressiveness is slowest, and specialness seems to
lie in between the two. This adds further important evidence regarding the processes
involved in aesthetic experiences and may also help to further develop current models of
aesthetic experience (e.g., Leder et al., 2004).

Acknowledgements
We would like to thank Katrien Torfs and Sven Panis for providing us with an algorithm to
construct our masks, Rudy Dekeerschieter for his IT support, and art historian Geert Verhavert
for helping us with the selection of stimuli. Thank you very much also to the Alte Pinakothek
Munich and the Yale University Art Gallery for granting the rights to reproduce artworks for
illustration purposes. This work was supported by long-term structural funding by the Flemish
Government (METH/08/02 and METH/14/02) to Johan Wagemans.

References
                           €
Allesch, C. (2006). Einfuhrung                              €
                                   in die psychologische Asthetik    [Introduction to psychological
    aesthetics]. Wien, Austria: Uni-Taschernb€ ucher, Wiener Universit€ats Verlag.
Artcyclopedia.com. Art cyclopedia: The fine art search engine [web database]. Calgary, Canada:
    Specifica, Inc. Retrieved from http://www.artcyclopedia.com/
Artstor. Artstor digital library [database]. New York, NY: Author. Retrieved from http://www.a
    rtstor.org/
Augustin, M. D., Carbon, C. C., & Wagemans, J. (2011a, August). Measuring aesthetic impressions of
    visual art (Poster). Presented at the 34th European Conference on Visual Perception (ECVP),
    Toulouse, France.
Augustin, M. D., Defranceschi, B., Fuchs, H. K., Carbon, C. C., & Hutzler, F. (2011b). The neural time
    course of art perception: An ERP study on the processing of style versus content in art.
    Neuropsychologia, 49, 2071–2081. https://doi.org/10.1016/j.neuropsychologia.2011.03.038
Augustin, M. D., Leder, H., Hutzler, F., & Carbon, C. C. (2008). Style follows content: On the
    microgenesis of art perception. Acta Psychologica, 128, 127–138. https://doi.org/10.1016/j.ac
    tpsy.2007.11.006
Augustin, M. D., & Wagemans, J. (2012). Empirical aesthetics, the beautiful challenge: An
    introduction to the special issue on Art & Perception. I-Perception, 3, 455–458. https://doi.org/
    10.1068/i0541aap
Augustin, M. D., Wagemans, J., & Carbon, C.-C. (2012). All is beautiful? Generality vs. specificity of
    word usage in visual aesthetics. Acta Psychologica, 139, 187–201. https://doi.org/10.1016/j.ac
    tpsy.2011.10.004
Bachmann, T. (2000). Microgenetic approach to the conscious mind. Amsterdam, The Netherlands:
    John Benjamins Publishing Company.
78   San Verhavert et al.

Bachmann, T., & Vipper, K. (1983). Perceptual rating of paintings from different artistic styles as a
    function of semantic differential scales and exposure time. Archiv Fur € Psychologie [Archive for
    Psychology], 135, 149–161.
Bar, M., Neta, M., & Linz, H. (2006). Very first impressions. Emotion, 6, 269–278. https://doi.org/10.
    1037/1528-3542.6.2.269
Berlyne, D. E. (1971). Aesthetics and psychobiology. New York, NY: Appleton-Century-Crofts.
Berlyne, D. E. (1974). Studies on the new experimental aesthetics. Steps toward an objective
    psychology of aesthetic appreciation. New York, NY: Wiley.
Brattico, E., Bogert, B., & Jacobsen, T. (2013). Toward a neural chronometry for the aesthetic
    experience of music. Frontiers in Psychology, 4, 1–21. https://doi.org/10.3389/fpsyg.2013.
    00206
Burke, E. (1990). A philosophical inquiry into the origin of our ideas of the sublime and beautiful
    (Vol. XXIX). Oxford, UK: Oxford University Press. (original work published 1757)
Cupchik, G. C., & Berlyne, D. E. (1979). The perception of collative properties in visual stimuli.
    Scandinavian Journal of Psychology, 20, 93–104. https://doi.org/10.1111/j.1467-9450.1979.
    tb00688.x
Cupchik, G. C., & Gebotys, R. (1988). The experience of time, pleasure, and interest during aesthetic
    episodes. Empirical Studies of the Arts, 6, 1–12. https://doi.org/10.2190/5yn3-j3p8-fwhy-udb3
Directmedia Publishing: zeno.org. (2007). 25000 Meisterwerke [25 000 masterpieces] [DVD-
    ROM]. Berlin, Germany: Directmedia Publishing.
Dodonov, Y. S., & Dodonova, Y. A. (2012). Response time analysis in cognitive tasks with increasing
    difficulty. Intelligence, 40, 379–394. https://doi.org/10.1016/j.intell.2012.07.002
Fei-Fei, L., Iyer, A., Koch, C., & Perona, P. (2007). What do we perceive in a glance of a real-world
    scene? Journal of Vision, 7(1), 1–10. https://doi.org/10.1167/7.1.10
Grill-Spector, K., & Kanwisher, N. (2005). Visual recognition: As soon as you know it is there, you
    know what it is. Psychological Science, 16, 152–160. https://doi.org/10.1111/j.0956-7976.
    2005.00796.x
Hegde, J. (2008). Time course of visual perception: Coarse-to-fine processing and beyond. Progress
    in Neurobiology, 84, 405–439. https://doi.org/10.1016/j.pneurobio.2007.09.001
Jacobsen, T. (2010). Beauty and the brain: Culture, history and individual differences in aesthetic
    appreciation. Journal of Anatomy, 216, 184–191. https://doi.org/10.1111/j.1469-7580.2009.
    01164.x
Jacobsen, T., Buchta, K., K€   ohler, M., & Schr€oger, E. (2004). The primacy of beauty in judging the
    aesthetics of objects. Psychological Reports, 94, 1253–1260. https://doi.org/10.2466/pr0.94.
    3c.1253-1260
Jacobsen, T., & H€   ofel, L. (2003). Descriptive and evaluative judgment processes: Behavioral and
    electrophysiological indices of processing symmetry and aesthetics. Cognitive, Affective, &
    Behavioral Neuroscience, 3, 289–299. https://doi.org/10.3758/cabn.3.4.289
Kant, I. (1991). Beobachtungen uber  €             € des Sch€
                                           das Gefuhl         onen und Erhabenen [Observations on
    the feeling of the beautifull and sublime]. Erlangen, Germany: Fischer. (original work published
    1764)
Knoop, C. A., Wagner, V., Jacobsen, T., & Menninghaus, W. (2016). Mapping the aesthetic space of
    literature “from below”. Poetics, 56, 35–49. https://doi.org/10.1016/j.poetic.2016.02.001
Leder, H., Belke, B., Oeberst, A., & Augustin, M. D. (2004). A model of aesthetic appreciation and
    aesthetic judgments. British Journal of Psychology, 95, 489–508. https://doi.org/10.1348/
    0007126042369811
Leder, H., & Nadal, M. (2014). Ten years of a model of aesthetic appreciation and aesthetic
    judgments: The aesthetic episode – Developments and challenges in empirical aesthetics.
    British Journal of Psychology, 105, 443–464. https://doi.org/10.1111/bjop.12084
Locher, P., Krupinski, E. A., Mello-Thoms, C., & Nodine, C. F. (2007). Visual interest in pictorial art
    during an aesthetic experience. Spatial Vision, 21, 55–77. https://doi.org/10.1163/156856
    807782753868
Beauty in the blink of an eye   79

Markovic, S. (2010). Aesthetic experience and the emotional content of paintings. Psihologija, 43,
   47–64. https://doi.org/10.2298/psi1001047m
Psychology Software Tools Inc. (n.d.). E-Prime 2.0 [Computer software].
Redies, C. (2015). Combining universal beauty and cultural context in a unifying model of visual
   aesthetic experience. Frontiers in Human Neuroscience, 9, 218. https://doi.org/10.3389/
   fnhum.2015.00218
Roye, A., H€ ofel, L., & Jacobsen, T. (2008). Aesthetics of faces. Journal of Psychophysiology, 22,
   41–57. https://doi.org/10.1027/0269-8803.22.1.41
Schopenhauer, A. (1948). Die Welt als Wille und Vorstellung [The world as will and
   representation] (Vol. 2 book 3). Z€   urich, Germany: Classen. (original work published 1818)
Smith, J. K., & Smith, L. F. (2001). Spending time on art. Empirical Studies of the Arts, 19, 229–236.
   https://doi.org/10.2190/5mqm-59jh-x21r-jn5j
Todorov, A., Pakrashi, M., & Oosterhof, N. N. (2009). Evaluating faces on trustworthiness after
   minimal time exposure. Social Cognition, 27, 813–833. https://doi.org/10.1521/soco.2009.27.
   6.813
Todorov, A., & Porter, J. M. (2014). Misleading first impressions different for different facial images of
   the same person. Psychological Science, 25, 1404–1417. https://doi.org/10.1177/09567976
   14532474
Torfs, K., Panis, S., Barthlema, A., & Wagemans, J. (2012, September). Temporal dynamics of
   candidate interpretations during object recognition. Presented at the 35th European Conference
   on Visual Perception (ECVP), Alghero, Italy.
Vernon, R. J., Sutherland, C. A., Young, A. W., & Hartley, T. (2014). Modeling first impressions from
   highly variable facial images. Proceedings of the National Academy of Sciences of the United
   States of America, 111, E3353–E3361. https://doi.org/10.1073/pnas.1409860111
Wagemans, J. (2011). Towards a new kind of experimental psycho-aesthetics? Reflections on the
   Parallellepipeda project. I-Perception, 2, 648–678. https://doi.org/10.1068/i0464aap

Received 6 May 2016; revised version received 24 May 2017

Appendix A: List of experimental stimuli per art style

Art style                     Painting                          Painter                         Year

Renaissance        The Alexander battle             Altdorfer, Albrecht                  1529
                    (Battle of Issus)
                   Portrait of an old Woman         Giorgione                            Ca. 1500–1510
                   Santa Maria Novella              Ghirlandaio, Domenico                1485–1490
Mannerism          The hell                         Beccafumi, Domenico                  1526–1530
                   The stone cutter                 van Hemessen, Jan                    Ca. 1514–1550
                                                     Sanders
                   Sheppard                         Bloemaert, Abraham                   1628
Baroque            Genre scene with Masks           Bonito, Guiseppe                     Mid-18th century
                   The stoning of Saint Stephan     Elsheimer, Adam                      Ca. 1600
                   Still life with fish             van Beyeren, Abraham                 1655–1666
                                                     Hendriksz
Classicism/        Cossack girl at                  Chasseriau, Theodore               1851
 Neoclassicism      Mazeppa’s body

                                                                                                  Continued
80   San Verhavert et al.

Appendix A (Continued)

Art style                     Painting                     Painter                    Year

                    Oedipus                    Ger^
                                                   ome, Jean-Leon             1867–1868
                    Self-portrait              Feuerbach, Anselm               1854
Rococo              Blind man’s bluff          Troost, Cornelis                2nd quarter
                                                                                18th century
                    The Strode family          Hogarth, William                1738
                    Head of a woman            Boucher, Francßois              Ca. 1750
Romantism           Honeymoon                  von Schwind, Moritz             1867
                    Daybreak                   von Schwind, Moritz             1858
                    Lake Tahoe                 Bierstadt, Albert               1868
Realism             Arrival of the sorcerers   Maximow, Wassilij               1875
                     at the wedding             Maximowitsch
                    Village politics           Leibl, Wilhelm Maria Hubertus   1877
                    Steel press                von Menzel, Adolf Friedrich     1872–1875
                     (modern day Cyclopes)      Erdmann
Impressionism       Pathway in a field         Degas, Edgar                    1890
                    Comtesse Adele de         Toulouse-Lautrec, Henri de      1881–1883
                     Toulouse-Lautrec,
                     at Breakfast
                    Glowing of the sun         Claus, Emile                    1905
Art Nouveau         Dance                      Mucha, Alfons                   1898
                    Three Woman and            Grasset, Eugene                1900
                     Three Wolves
                    The Autumn Bride           Levy-Dhurmer, Lucien            1896
Expressionism       The World-cow              Marc, Franz                     1913
                    The Windbride              Kokoschka, Oskar                1913
                    Carnival in Arcueil        Feininger, Lyonel               1911
Cubism              Denstedt                   Feininger, Lyonel               1917
                    Femme Assise               Metzinger, Jean                 1919
                    Still-life with scull      Kubista, Bohumil                1912
Abstract            After Khorkum              Gorky, Arshile                  1940–1942
 expressionism
                    Untitled                   Gorky, Arshile                  1943
                    Arabesque                  Hamilton Bush, Jack             1975
Surrealism          Fascination                Brauner, Victor                 1939
                    How to make a rainbow      Cornell, Joseph                 ca. 1963, 1965
                    Dutch Interior II          Miro, Joan                    ?
Action painting     Minter                     Poons, Larry                    1975
                    White squares              Krasner, Lee                    ca. 1948
                    Number 7                   Pollock, Jackson                1951
Pop Art             The continuous monument    Natalini, Adolfo; di            1969
                                                 Francia, Cristiano
                                                 Toraldo; Superstudio;
                                                 Magris, Alassandro;
                                                 Frassinelli, Gian Piero
                    Alka Seltzer               Lichtenstein, Roy               1966
                    Painted Bronze             Johns, Jasper                   1960
Photorealism/       Water                      Gertsch, Franz                  ?
 Hyperrealism

                                                                                       Continued
Beauty in the blink of an eye   81

Appendix A (Continued)

Art style                   Painting                          Painter                          Year

                 Trace Chain II                   Don Eddy                             1995
                 Waverly Place                    Estes, Richard                       1980
Fauvism          The Piano Lesson                 Matisse, Henri                       1916
                 London Bridge                    Derain, Andre                       1906
                 Woman at the Window              Friesz, Othon                        1919
Concrete Art     Thirty vertical systematic       Lohse, Richard Paul                  1970
                  colour series in a yellow
                  rhombic form
                 Tekers-MC                        Vasarely, Victor                     1981
                 Two cells with conduit           Halley, Peter                        1987

Appendix B: List of practice stimuli per art style

Art style                              Painting                  Painter                      Year

Renaissance                 Annunciation                  Albertinelli, Mariotto    1508
Mannerism                   Farmers carnival              Bol, Hans                 2nd half 16N century
Baroque                     An angel opens the            Cuyp, Benjamin            1640
                             grave of Christ               Gerritsz
Classicism/                 Aeneas reports Dido           Guerin,                  1815
 Neoclassicism               about the downfall            Pierre-Narcisse
                             of Troy
Rococo                      The chemist                   Longhi, Pietro            Ca. 1752
Romantism                   Canal Grande in Venice        Turner, Joseph            1835
                                                           Mallord William
Realism                     Fox in the snow               Courbet, Gustave          1860
Impressionism               Road to Port-Marly            Pisarro, Camille          1860–1867
Art Nouveau                 Love                          Klimt, Gustav             1895
Expressionism               The Vampire                   Munch, Edvard             1895
Cubism                      Couple                        Pascin, Jules             1915
Abstract expressionism      Carnival                      Gorky, Arshile            1943
Surrealism                  The great Sirens              Delvaux, Paul             1947
Action Painting/abstract    Cotherman                     Poons, Larry              1981
 impressionism
Pop Art                     Landscape 3                   Lichtenstein, Roy         1967
Photorealism/Hyperrealism   Luciano Castelli I            Gertsch, Franz            1971
Fauvism                     Interior with seated Figure   Matisse, Henri            1920–1921
Concrete Art                Variation 13                  Bill, Max                 1938
82   San Verhavert et al.

Appendix C: Results simple main effects analysis on correlations
Table C1. Results of simple main effects analysis on Fisher’s Z corrected correlations for
Experiment 1

Comparison                               M                     SE                        95% CIa

PT
  10–50 ms                              .242**                 .052                 [   .347;   .137]
  10–100 ms                             .388**                 .062                 [   .512;   .263]
  10–500 ms                             .511**                 .064                 [   ,640;   .382]
  50–100 ms                             .145*                  .050                 [   .246;   .045]
  50–500 ms                             .268**                 .053                 [   .374;   .162]
  100–500 ms                            .123*                  .045                 [   .213;   .032]
Scale
  Impressive–beautiful                  .285**                 .040                 [ .365; .206]
  Impressive–special                    .110*                  .038                 [ .187; .033]
  Beautiful–special                     .175**                 .034                 [ .107;.243]

Note. M = mean difference; SE = standard error; 95% CI = 95% confidence interval.
*p < .01; **p < .001.
a
 CI Least significance difference corrected.

Table C2. Results of simple main effects analysis on Fisher’s Z corrected correlations for
Experiment 2

Comparison                               M                     SE                        95% CIa

PT
  20–30 ms                              .167**                 .030                 [ .227; .108]
  20–40 ms                              .190**                 .034                 [ .258; .122]
  30–40 ms                              .023                   .032                 [ .078;.033]
Scale
  Impressive–beautiful                  .217**                 .056                 [ .329; .105]
  Impressive–special                    .097*                  .045                 [ .188; .007]
  Beautiful–special                     .120*                  .057                 [.005;.234]

Note. M = mean difference; SE = standard error; 95% CI = 95% confidence interval.
*p < .05; **p < .001.
a
 CI Least significance difference corrected.
You can also read